VERBAL AUTOPSIES FOR ADULT DEATHS - THEIR DEVELOPMENT AND VALIDATION IN A MULTICENTER STUDY

BACKGROUND Verbal autopsy (VA) has been widely used to ascertain cause s of child deaths, but little is known about the usefulness of VA for adult deaths. This paper describes the process used to develop a VA to ol for adult deaths and the results of a multicentre validation of thi s tool. METHODS A mortality classification was developed by including causes of death that might be arrived at by VAs and causes that are re sponsive to public health interventions. An algorithm was designed for each cause in the classification, based on classifying symptoms into essential, supportive and differential. A structured questionnaire des igned to elicit information on these symptoms was developed in English translated into the local languages. The tool was validated on deaths occurring at hospitals in Tanzania (315 deaths), Ethiopia (249) and G hana (232). Hospital records of all adult deaths occurring at the stud y hospitals from June 1993 to April 1995 were collected prospectively Non-medical interviewers with at least 12 years of formal education co nducted VA interviews. Causes of death were diagnosed by a panel ol: p hysicians and by a computerized algorithm. The validity of the VA was assessed by comparing the VA diagnoses with hospital diagnoses. RESULT S Specificity of VAs by physicians fell below 95% only for acute febri le illness (AFI) and TB/AIDS. Sensitivity and positive predictive valu e (PPV), however, varied widely both across the sites and between caus es. Sensitivity was > 75% for tetanus, rabies, direct maternal causes, injuries and TB/AIDS and ranged between 60% and 74% for diarrhoea, ac ute abdominal conditions and AFI. The PPV was > 75% for tetanus, rabie s, hepatitis and injuries and ranged between 60 and 74% for meningitis , AFI, TB/AIDS and direct maternal causes. When the communicable disea ses were combined in a single group, the sensitivity was 82%, specific ity 78% and PPV 85%. For the group of noncommunicable: diseases the co rresponding sensitivity specificity and PPV were 71%, 87% and 67%, res pectively Use of all algorithm resulted in lower sensitivity, specific ity and PPV than the VAs by physician. CONCLUSIONS VAs by a panel of p hysicians performed better than all opinion-based algorithm. The valid ity of VA diagnosis was highest for AFI, direct material causes, TB!AI DS, tetanus, rabies and injuries.